\begin{tabular}{l cc c cc c cc c cc c }
\hline \hline
 Dependent variable: &  \multicolumn{2}{c}{College quality index} && \multicolumn{2}{c}{College's 75th pctile math SAT score} && \multicolumn{2}{c}{STEM major} && \multicolumn{2}{c}{Graduate degree}  \\
 \cline{2-3} \cline{5-6} \cline{8-9} \cline{11-12} 
 &  (1) & (2) && (3) & (4) && (5) & (6) && (7) & (8)  \\ 
 \cline{2-3} \cline{5-6} \cline{8-9} \cline{11-12} 
\\ \multicolumn{8}{l}{\emph{Panel A: Independent variables are binary measures of over- and under-confidence}} \\ \\
\hspace{10pt} Over-confidence &  ---  &  ---  &&  ---  &  ---  &&  0.067  &  0.076  &&  0.035  &  0.032 \\
  &    &    &&    &    &&  (0.096)  &  (0.097)  &&  (0.083)  &  (0.087) \\
\hspace{10pt} Under-confidence &  ---  &  ---  &&  ---  &  ---  &&  -0.158$^{***}$  &  -0.162$^{***}$  &&  0.014  &  0.006 \\
  &    &    &&    &    &&  (0.036)  &  (0.036)  &&  (0.046)  &  (0.048) \\
\hspace{20pt} N & ---  &  ---  &&  ---  &  ---  &&  736  &  736  &&  810  &  810 \\
\\
\hspace{20pt} OC = -1*UC?  \emph{p-value}:  & ---  &  ---  &&  ---  &  ---  &&  0.365  &  0.405  &&  0.607  &  0.704 \\
\\ \multicolumn{8}{l}{\emph{Panel B: Independent variable is degrees of over- and under-confidence in standard deviation units}} \\ \\
\hspace{10pt} Confidence &  ---  &  ---  &&  ---  &  ---  &&  0.077$^{***}$  &  0.078$^{***}$  &&  0.023  &  0.022 \\
  &    &    &&    &    &&  (0.023)  &  (0.023)  &&  (0.025)  &  (0.025) \\
\hspace{20pt} N & ---  &  ---  &&  ---  &  ---  &&  736  &  736  &&  810  &  810 \\
\\  Sample mean of dep. var. &  &&&  &&& 0.189 &&& 0.200 & \\ \\
Basic controls: & \checkmark & \checkmark && \checkmark & \checkmark && \checkmark & \checkmark && \checkmark & \checkmark   \\
Added background controls: &                     & \checkmark &&                     & \checkmark &&                     & \checkmark &&                     & \checkmark   \\
\hline \hline 
\end{tabular}
\begin{tabular}{ p{7in} }
\footnotesize
Notes:  This table regresses college outcomes outcomes on childhood biased beliefs with various controls.          Biased beliefs are measured in the earlies observed wave in the CDS with non-missing test scores and self-assessed ability.          In Panel A, the outcome is regressed on an indicator for over-confidence, an indicator for under-confidence and our basic set of controls (in odd-numbered columns) and our extended set of controls (in even-numbered columns). The p-value listed tests whether the coefficient on the over-confidence indicator is equal to -1 times the coefficient on the under-confidence indicator. In Panel B, the outcome is regressed on our more continous measure of biased beliefs which has been standardized to have mean zero and standard deviation one in our sample and the same sets of controls.          All controls are the same as described in Table \ref{tab:persist_conf}, minus the controls for adolescent test score deciles.                       Standard errors are clustered at the family level and included in parentheses below each          estimate. *, **, and *** indicate significance at the 0.1, 0.05, and 0.01 percent level, respectively. 
\end{tabular}
